STUDENTS ’ VOICES ON LECTURERS ’ WRITTEN CORRECTIVE FEEDBACK ( WCF ) IN THEIR WRITING TASKS

ABSTRACTAlthough studies on Written Corrective Feedback (WCF, hereafter) have been increasingly prevalent in the last few years, inquiries on how advisory students perceive the lecturers’ feedback on their writing tasks have been likely scarce, especially in Indonesian Higher Education Institutions (HEIs) contexts. This study examines the students’ perception and evaluation of the lecturers’ WCF in response to errors and inaccuracies in their academic writing tasks. Through an online survey questionnaire distributed to 46 respondents via email, the results show that the majority of students appreciated any forms of feedback from the lecturers. Their writing skills in four aspects (grammar, vocabulary, organization, and mechanics) also improved significantly through an enhancement of their self-directed learning. Following the analysis model by one of previous studies, the results showed that the students preferred direct WCF to the Indirect one (58.7 %: 15.2 %), while the “Praise” category was given the highest rate with an average score of (4.06). “Criticism”, on the other hand, was the lowest one with an average score of only (2.3) in the evaluation. It is recommended that lecturers always avoid unclear, vague, aggressive, thoughtless, and inappropriate feedback to improve students’ writing skills and performance. ABSTRAKMeskipun studi tentang Umpan Balik Korektif Tertulis (WCF, selanjutnya) semakin lazim dalam beberapa tahun terakhir, namun, diskusi tentang penilaian mahasiswa terhadap umpan balik dosen terkait tugas menulis mereka sangat terbatas, terutama dalam konteks Lembaga Pendidikan Tinggi Indonesia (PT). Penelitian ini mengkaji persepsi dan evaluasi mahasiswa terhadap WCF dosen dalam merespon kesalahan dan ketidaktepatan dalam tugas menulis akademik mahasiswa. Melalui kuesioner survei online yang dibagikan kepada 46 responden melalui email, hasilnya menunjukkan bahwa mayoritas mahasiswa mengapresiasi segala bentuk umpan balik dari dosen. Keterampilan menulis mereka dalam empat aspek (tata bahasa, kosakata, organisasi, dan mekanik) juga meningkat secara signifikan melalui peningkatan pembelajaran mandiri mereka. Mengacu kepada model analysis dari salah satu penelitian terdahulu, ditemukan bahwa penggunaan WCF Langsung lebih disukai oleh siswa daripada penggunaan Tidak Langsung (58,7%: 15,2%), sedangkan kategori “Pujian” diberikan nilai tertinggi dengan nilai rata-rata (4,06). Sementara, “Kritik” adalah yang paling rendah dengan skor rata-rata hanya (2,3) dalam evaluasi. Disarankan agar dosen selalu memberikan semangat, namun hindari umpan balik yang tidak jelas, kabur, agresif, ceroboh, dan tidak tepat untuk meningkatkan kemampuan menulis dan prestasi mahasiswa.


INTRODUCTION
There has been a growing interest in studying written corrective feedback (WCF) provided by lecturers on students' writing tasks in the last several years (Bitchener, 2008;Bitchener & Knoch, 2009;Kumar & Stracke, 2007;Adrefiza & Fortunasari, 2020). Although its efficacy and effectiveness remain controversial (Truscott, 1996;Bitchener, 2008), studies have suggested that WCF plays a significant role in developing students' self-regulatory learning (Stracke and Kumar, 2010) and significantly accelerates the students' writing skills and performance. WCF has also been claimed to boost students' autonomous learning in writing (Adrefiza & Fortunasari, 2020). It strengthens and facilitates students' self-regulatory learning through an engagement process in which they become accustomed to handling their errors and inaccuracies from lecturers' feedback.
Such an essential improvement in the students' skills may be regarded as more effective than that obtained from the formal teaching of writing itself.
WCF is also believed to support and maintain the advisory students' psychological and personal relationships and the lecturers (Kumar & Stracke, 2007). This happens simply because both students and supervisors are actively engaged in distant communication through queries, clarifications, suggestions, and instructions over the students' writing problems and inaccuracies. Inappropriate feedback and responses within the communication, to some extent, may result in student's discouragement and even a personal conflict between the two parties, causing students' neglection and disappointment, especially when the feedback and responses are not carefully and wisely addressed. Unclear, vague, ambiguous, inappropriate forms, aggressive, impulsive, and ineffective WCF provisions from the lecturers may depress and demotivate students in many ways, and this is possible to account for adverse effects on students' learning.
This study attempts to look at students' perceptions and evaluation of the lecturers' WCF and see how it should be carefully provided and handled by the lecturers. The negative impacts on students' learning can be thoughtfully anticipated and minimised. These research results are expected to provide a critical evaluation of student-lecturer supervisory practices at local Indonesian HEIs and hopefully contribute to an appropriate, effective, and workable method in assessing the lecturers' WCF in the Studies on WCF provided by lecturers on students' writing tasks have been increasingly predominant in the last several years (Kang & Han, 2015;Storch, 2010). Although studies, such as those from Kumar and Stracke (2007), Hyland and Hyland (2006), and Stracke and Kumar (2010) confirm that lecturers' WCF encourages students' self-directed learning (SRL), especially in improving second or foreign language learners' writing skills. However, Storch (2010) claims that the findings are still unsettled and suggest further robust directions and practical studies to address their shortcomings. This is likely to be evident since the results tend to be controversial and uncertain. Bitchener (2008) claims that studies on WCF have long been debatable due to their controversial design and efficacy, pointing to an issue that they often produce conflicting results which are, in fact, potentially affected by various aspects, such as the means of research methods used in the studies and the contexts in which the studies are conducted.
WCF is often linked to SRL, known as a platform of academic learning which gained prominence in the mid of 1980s (Stracke & Kumar, 2010;Kang & Han, 2015). This kind of learning gives rise to individual attempts, based on their concentration and interest, to coordinate and manage their learning. Zimmermann (2001) argues that in this mechanism, the individual students set up their own learning experience and strategies that fit their goals through active metacognitive, motivation, and behaviour to self-generate thoughts, feelings, and actions to achieve their proposed and predetermined learning goals. This active involvement is said to be an essential characteristic of SRL (Zimmermann, 2001in Stracke & Kumar, 2010, that is why SRL is also seen as a contributing factor in successful learning and academic achievement as students independently determine what to do and learn from other people without any external influences. According to Zimmermann (2001), through SRL, the students develop a dynamic process and transferability to keep moving and never stand static. This type of learning is believed to go in line with the context of thesis supervisory practices or writing supervisions at HEIs, in which the student is required to seek information actively and perform necessary tasks to address all WCF addressed by the supervisors, and this certainly requires extra efforts and time to do (Boekaerts, 1999). In this platform, SRL serves a function as self-directed 13-23 http://journal.uinjkt.ac.id/index.php/ijee | DOI: http://doi.org/10.15408/ijee.v8i1.17701 P-ISSN: 2356-1777, E-ISSN: 2443-0390 | This is an open access article under CC-BY-SA license learning (SDL) for which WCF plays a role as an indispensable substance (Butler & Winne, 1995) and to university students, according to Stracke and Kumar (2010), WCF in this context, lies at the heart of the SRL experience for them.
The practice of SRL by providing WCF in the students' academic writing is likely uncommon in most local Indonesian HEIs (Adrefiza & Fortunasari, 2020). They claim that studies on how WCF affects students' learning and motivation have been almost absent in academic atmospheres and students rarely receive enough and meaningful feedback from the lecturers to enhance their SRL. At the same time, related studies have often been mainly focused and world-widely on Western ESL or EFL contexts. This has resulted in an unbalanced concentration of the investigation, causing an extensive gap in WCF studies in Western and Eastern contexts. As a result, there is still a limited investigation of WCF in the countries where English is studied either as a second or foreign language exists. Simultaneously, the absence of WCF in the students' academic writing tasks makes the students' learning ineffective. Students do not learn much from the lecturers' feedback. Neither do they understand their errors and inaccuracies in their writing as the communication with the lectures is near to absence?
With its SRL nuance, WCF encourages collaboration and personal networking between both students and supervisors. Mullins andKiley (1998, as cited in Kumar &Stracke, 2007) claim that communication and collaboration are practically necessary for university student-supervisory practices. Without intensive care of these two aspects, the students cannot optimally improve their writing skills, nor do they effectively develop their academic and interpersonal communication with the lecturers. It is further stressed that these two generic skills are considered imperative and workable in the student academic writing supervisions. They are believed to play a vital role in the students' educational development process and even ultimately serve a notable function as professional dexterity in Higher Education outcomes (Philips & Pugh, 2005).
WCF is seen as a reflection of the lecturers' feedback and instructions about the students' work, guiding them to specific points that they need to address and explain. To make appropriate changes to the lecturers' WCF, learners need to take serious care and attention. Overall, WCF and students' responses form a personal communication between students and 14-23 http://journal.uinjkt.ac.id/index.php/ijee | DOI: http://doi.org/10.15408/ijee.v8i1.17701 P-ISSN: 2356-1777, E-ISSN: 2443-0390 | This is an open access article under CC-BY-SA license lecturers, promoting self-regulated learning (Stracke & Kumar, 2010) as the interaction facilitates critical thought, study, and exploration critical in the academic growth of students in higher education. Nevertheless, in many local English Education Programs in Indonesian Tertiary Education, such a mechanism is probably not apparent.
WCF illustrates the communication between lecturer and student as though the lecturer communicates his thoughts and ideas in a face-to-face atmosphere regarding the student's writing duties. It also reveals the lecturer's emotions and thoughts, expressed in the content and quality of the student's writing. WCF is also expressed by unique written symbols such as interjection (!), question mark (?), and emoticons, unlike in actual face-to-face interaction. It mimics and other non-verbal gestures. Frequently, by the repeating number of symbols used, the force of the emotion is shown. The more symbols are used in WCF, the more significant or more serious the writer's concern. To sum up, WCF's use illustrates the teacher or lecturer's personal and psychological status over the writing mistakes and inaccuracies of the student.
Holmes' (2008) model of speech act function classification has been used in a few WCF studies. Kumar and Stracke (2007) and Stracke and Kumar (2010), for example, bring this categorisation to investigate WCF's provision in its relation to this classification. Their studies categorise WCF into three main speech act categories (Referential, Directive, and Expressive). Subcategories for each main category are provided in the table 1.

Organisation
This does not belong in the literature review.

Content
Are you sure you can make such a claim?

Directive Suggestion
Maybe this is not necessary.

Question
Whose term is this?
Instruction Please clarify.

Expressive Praise
Good, nice example.

Criticism
This table…does not add to the text.

I would be interested in exploring what
triggered this (Kumar & Stracke, 2007: 464) Ellis (2010) proposes several ways in which WCF can be formulated. In general, WCF can be expressed in two basic types, direct and indirect. The former is usually represented by providing the correct forms over the errors or inaccuracies before marking the incorrect forms with other remarks such as underlines, question marks,

15-23
http://journal.uinjkt.ac.id/index.php/ijee | DOI: http://doi.org/10.15408/ijee.v8i1.17701 P-ISSN: 2356-1777, E-ISSN: 2443-0390 | This is an open access article under CC-BY-SA license crosses, removals, and symbols. On the other hand, the latter is often formulated through various markings such as queries, questions, asking for confirmation, directions, and many other forms of expressions both in the forms of linguistic expressions and nonlinguistic features. According to Ellis (2010), metalinguistic feedback is also prominent and frequent in the lecturers' WCF. This type of feedback is usually represented through prompts and cues to indicate the students' writing errors and inaccuracies. Students are required to analyse the meanings and directions of the feedback so that they need to respond appropriately and accordingly.

See an example:
Example of Direct WCF: A dog stole (a) bone from (a) butcher. He escaped with having (the) bone. When the dog was going (over)through the bridge over (a) the river, he (saw a) found dog in the river.
In the above example, the feedback is noted by crossing out the incorrect or unnecessary words, phrases, or morphemes. It may be done by inserting a missing or expected word and providing the correct linguistic form above or near the error. According to Ellis (2009), such a type benefits students effectively as it delivers explicit information about the right answer.
Indirect feedback is another type of WCF provided by lecturers in which the correct forms are not written directly. Instead, the lecturers indicate or mark the errors' location (Ferris & Roberts, 2001). This type is intended to make the students aware that they have an error in their writing. The errors can be noted using a few symbols, such as an underline, a circle, a code, a mark, or a highlight. Here is an example below: Example of Indirect WCF: A dog stole X bone from X butcher. He escaped with XhavingX Xbone. When the dog was going XthroughX X bridge over XtheX river, he found X dog in the river.
X= missing word X__X = wrong word Apart from its weak advantages, this type of WCF is said to have a few strengths. According to Lalande (1982), the students tend to be more alert and creative in responding and making the corrections, as they get themselves engaged actively so that they learn with a good reflection, which can lead to long-term memory. However, some researchers argue that indirect corrective feedback is less beneficial to lower proficiency levels because they lack the meta-linguistic awareness

16-23
http://journal.uinjkt.ac.id/index.php/ijee | DOI: http://doi.org/10.15408/ijee.v8i1.17701 P-ISSN: 2356-1777, E-ISSN: 2443-0390 | This is an open access article under CC-BY-SA license needed to correct their mistake (Ferris, 2004;Hyland, 2006). This research investigates the students' perception and evaluation of the WCF provided by the lecturers in their writing tasks. The results are expected to be useful for better WCF provisions by lecturers in responding to the students' errors and inaccuracies in their academic writing tasks to improve the students' self-directed learning and, in turn, increase the students' writing skills and performance. Three research questions were posed: (1) What type of WCF is best preferred by the students?
(2) How do the students perceive the lecturers' WCF on their academic writing tasks? Moreover, (3) How do the students learn from the WCF?

METHOD
This is a quantitative study with descriptive analysis, aiming at exploring students' perception and evaluation of lecturers' WCF provided on their academic writing tasks at English Study Program, the Faculty of Education, Jambi University. It involved 46 randomly selected students (23 males and 23 females), both from S1 (Undergraduate) and S2 (Postgraduate) levels, who were in their final semester and the process of thesis supervision.
The data were gathered through a closed-ended questionnaire (using a Likert scale model), sent via email. The respondents were required to provide their evaluation and evaluation on the WCF provided by the lecturers on their academic writing tasks. The questionnaire comprises some items which embrace the students' responses on four main issues: (a) students' preferences on types of lecturers' WCF -direct versus indirect; (b) students' evaluation on WCF types according to speech acts functions -editorial, referential, and expressive; (c) students' evaluation on the use of non-linguistic features on the lecturers' WCF; and (d) students' writing improvements through the lecturers' WCF. Following Kumar and Stracke (2007) and Stracke and Kumar (2010), the responses were tabulated and categorised based on their rates and categorisations.

Students' preference for direct and indirect WCF
Students' preference for the lecturers' WCF shows an interesting phenomenon. As shown in Table 2 below, the disparity between two continuums, Direct and Indirect, is relatively high. Many of the students prefer the direct to the indirect type. See the following   Table 2 shows the students' preference for lecturers' WCF. More than 50 % of the students prefer direct to indirect type (27:7) or (58.7 %: 15.2 %). There are twelve (26 %) students who like the combination of both direct and indirect types. This preference might be based on the students' experience in dealing with lecturers on their writing tasks, especially in their attempts to respond and understand the real messages or corrections that the lecturers try to deliver in their WCF. In direct WCF, the lecturers usually provide direct correction or replacement over the students' errors or mistakes. Simultaneously, in the indirect one, the correct forms are not given but only addressed with clues or prompt that guide the students to the correct forms. The distribution reflects that direct WCF was preferred most by the students, and they were not happy with the indirect one. Table 3 shows the students' evaluation of the types of WCF provided in their academic writing. As all WCF is classified into nine categories based on three main categories of speech act functions (referential, directive, and expressive), the evaluation seems varied according to each type. The scores' distribution looks disperse, showing an unpredictable trend, as shown in Table 3. Regarding the detailed types on WCF according to speech act classifications, as shown in Table 3, the distribution shows an interesting phenomenon. The highest rate is given to "Praise" with an average score (4.06), while "Criticism" receives the lowest proportion with only (2.3) on average. At the same time, "Suggestion", "Opinion", and "Editorial" remain relatively high, reaching over 3. Other types that receive a relatively low evaluation from the students are "Instruction", "Organization", "Content", and "Question". These categories are rated lower than 3.0 on average in the evaluation.

Students' evaluation on the use of non-linguistic features in the lecturers' WCF
Students' evaluation of the use of non-linguistic features on the lecturers' WCF shows an interesting phenomenon. As shown in Table 4 below, three types of features are dispersedly scored. Regarding the use of non-linguistic features in the lecturers' WCF, it appears that the use of arrows, circles, underlines, cross, ticks, and these sorts of things are mostly preferred with the average rate of 3.04. The other two categories (question mark and interjection) are not highly appreciated and rated only at 2.24 and 2.32.

Students' Evaluation on their Writing Improvement from the Lecturers' WCF
Students writing improvement lies in four aspects as the result of the lecturers' WCF. As shown in Table 5 below, the four aspects are scored dispersedly. As shown in Table 5, through the lecturers' WCF, students' writing improves in four areas (grammar, vocabulary, organisation, and mechanics). Grammar and vocabulary are the two rated high in the students' evaluation concerning the improvement of the lecturers' WCF. It was evident in the table that the two skills received a high rate with the average scores of (3.43) for both. "Organisation", on the other hand, the other two skills http://journal.uinjkt.ac.id/index.php/ijee | DOI: http://doi.org/10.15408/ijee.v8i1.17701 P-ISSN: 2356-1777, E-ISSN: 2443-0390 | This is an open access article under CC-BY-SA license (Organization and Mechanics) were rated low, with only 2.97 and 3.17 on the students' evaluation, respectively.

The students' appreciation of the lecturers' WCF
It is evident in the study that the students appreciated any form of feedback provided by the lecturers in their academic writing tasks. Stracke and Kumar (2010) suggested that WCF encourages students' self-regulatory learning since they are forced to respond to any queries, questions, and corrections from the lecturers. The feedback directs the students to revise their writing for improvement, which is usually undetectable for many students. The students often put the corrections in their memory to become alert in the next writing, which develops and accelerates the students' writing skills and performance. Through WCF, students are aware of their inaccuracies and errors, which increases students' autonomous learning in writing (Adrefiza & Fortunasari, 2020). Students learn a lot from the WCF since it forms a process in which they become accustomed to handling their errors and inaccuracies from the lecturers' feedback. This is often believed to more effective compared with that obtained from the formal teaching of writing itself.
Another essential benefit of WCF is maintaining the advisory students' psychological and personal relationships and the lecturers. This is supported by Kumar and Stracke (2007) since WCF enhances an active engagement between students and the lecturers through a distant communication practice. However, such communication may lead to conflicts, especially when the feedback and responses are not appropriately and wisely addressed. Some students may be sensitive and become discouraged from the lecturers' WCF. Hence, vague, ambiguous, inappropriate forms, aggressive, and impulsive WCF may be discouraging and depress the students in many ways so that they should be avoided.

The Students' Preferences on the Lecturers' WCF Types
The fact that many students prefer direct to indirect feedback is interesting to note. The results, at least, unfold two apparent phenomena. First, through direct WCF, the students get direct corrections from the lecturers either through direct replacements or suggested changes over the incorrect forms in the students' writing. Students do not have to spend extra time, though, and energy to revise the writing with this type, as every correction has been made clear to them. Second, such a feedback model does not cause any psychological and mental burden to students since the response to the inaccuracies can be done effortlessly. This does not work for the indirect type. Students will have to interpret the feedback before they revise the incorrect forms criticallyand sometimes this is time consuming and discouraging. This is so as Ellis (2010) states that the indirect WCF is often formulated through various markings such as queries, questions, asking for confirmation, directions, and many other forms of expressions both in the forms of linguistic expressions and non-linguistic features. According to Ellis (2010), this type of feedback frequently uses metalinguistic features that are usually represented through prompts and cues to indicate the errors and inaccuracies in the students' writing.
It is also apparent that students best prefer WCF categories of "Praise" and "Suggestion". This is interesting because it reflects students' primary character, where the two positive rewards are highly appreciated. The students perceive them as motivating and encouraging feedback, but they usually perceive them as a psychological release and relaxation. Simultaneously, categories of "Criticism" and "Instruction" come out as the least preferred type of WCF among the respondents. These two categories may be perceived as a negative reward and bring a high psychological burden to them. With these two types of feedback, students will have to work hard to respond and revise the writing, resulting in a psychological burden. This is in line with what Kumar and Stracke (2007) suggest that WCF links to both the advisory students and the lecturers' psychological state and personal relationship.

CONCLUSION AND SUGGESTION
In general, the students have a positive perception and evaluation of the lecturers' WCF since it improves their writing skills in several ways. They appreciate all forms of WCF provided by the lecturers on their academic writing tasks. The highest rate of the responses indicates this addressed to positive categories in the questionnaires. The students claim that with lecturers' feedback, students learn from their errors and inaccuracies in their writing and anticipate the same errors in the future. To the same extent, this practice is believed to be more effective than formal learning of the writing itself.
The majority of the students prefer a direct type of WCF from the lecturers rather than the indirect one. They find it much easier to analyse and respond to direct feedback simply because they do not need to have rigid interpretations over the meaning and direction of the feedback as they usually do with the indirect type. The categories of "Praise", "Suggestion", and "Opinion" are the three types of WCF that are more preferred by many of the students. These categories were highly rated by the students. Other categories, such as: "Criticism", "Instruction", and "Organization", are less preferred to students as they put higher pressure on them.
Overall, through the lecturers' WCF, students learn and improve their writing in four main skills. They cover: (a) grammar; (b) vocabulary; (c) organization of ideas; and (d) mechanicals. Among the four aspects, students find that grammar was the highest one that improves significantly from the lecturers' WCF, followed by vocabulary, organisation of ideas, and mechanicals. The students learned independently and autonomously much more through lecturers' WCF than through other practice, including formal learning in the classroom. Students' errors should be wisely and carefully treated through effective WCF so that students' self-regulated learning is enhanced, and the students are not discouraged.